The goal of this proposal is to address the growing problem of prescription opioid abuse by examining the effects of oxycodone in a laboratory setting in the presence of clinical pain and following significant pain relief following maintenance on the buprenorphine / naloxone combination (Bup/NX). Participants diagnosed with moderate pain initially will be admitted to hospital and transitioned from their baseline prescription opioid to a standing daily dose of Bup/NX. During the Bup/NX maintenance periods, participants will have the opportunity to self-administer oxycodone and subjective responses to oxycodone as well as analgesic, physiological and performance effects will be measured. In the second phase of this study, the same patients who participated in the inpatient phase will be followed on an outpatient basis while maintained on Bup/NX. Primary outcome variables during the outpatient phase will include percentage of negative urine toxicology samples for opioids and number of weeks in treatment. A major goal of this proposal is to determine which variables collected in the laboratory most reliably predict subsequent relapse to opioid abuse. In addition, the utility of Bup/NX in treating patients diagnosed with both chronic pain and substance abuse will be assessed. Despite the significant increase in prescription opioid abuse over the past 5-10 years, very few studies have systematically examined the issues surrounding prescription opioid abuse in the population of individuals with chronic pain. To our knowledge, no studies to date have examined opioid self-administration in persons with pain who have a history of opioid abuse. Specifically, the current proposal aims to 1) characterize the behavioral effects of oxycodone in Bup/NX-maintained pain patients who are currently abusing their prescription opioids;and 2) examine the relationship between behavioral effects of oxycodone as measured in the laboratory setting, and subsequent relapse to opioid abuse. These studies will provide important information about prescription opioid abuse liability in pain patients and will yield a laboratory model for predicting likelihood to relapse to prescription opioid abuse. These questions are of immediate clinical relevance to the treatment of chronic pain with opioid therapy.